New scaling model for variables and increments with heavy-tailed distributions

Monica Riva, Shlomo P Neuman, Alberto Guadagnini

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Many hydrological (as well as diverse earth, environmental, ecological, biological, physical, social, financial and other) variables, Y, exhibit frequency distributions that are difficult to reconcile with those of their spatial or temporal increments, ΔY. Whereas distributions of Y (or its logarithm) are at times slightly asymmetric with relatively mild peaks and tails, those of ΔY tend to be symmetric with peaks that grow sharper, and tails that become heavier, as the separation distance (lag) between pairs of Y values decreases. No statistical model known to us captures these behaviors of Y and ΔY in a unified and consistent manner. We propose a new, generalized sub-Gaussian model that does so. We derive analytical expressions for probability distribution functions (pdfs) of Y and ΔY as well as corresponding lead statistical moments. In our model the peak and tails of the ΔY pdf scale with lag in line with observed behavior. The model allows one to estimate, accurately and efficiently, all relevant parameters by analyzing jointly sample moments of Y and ΔY. We illustrate key features of our new model and method of inference on synthetically generated samples and neutron porosity data from a deep borehole. Key Points: A new statistical scaling model is developed, explored and applied Apparent inconsistency between variable and increment statistics is eliminated Parameters are estimated using variable and increment moments jointly

Original languageEnglish (US)
Pages (from-to)4623-4634
Number of pages12
JournalWater Resources Research
Volume51
Issue number6
DOIs
StatePublished - Jun 1 2015

Fingerprint

distribution
borehole
porosity
parameter
method
statistics

Keywords

  • heavy-tailed distributions
  • neutron porosity
  • parameter estimation
  • scaling
  • sub-Gaussian

ASJC Scopus subject areas

  • Water Science and Technology

Cite this

New scaling model for variables and increments with heavy-tailed distributions. / Riva, Monica; Neuman, Shlomo P; Guadagnini, Alberto.

In: Water Resources Research, Vol. 51, No. 6, 01.06.2015, p. 4623-4634.

Research output: Contribution to journalArticle

Riva, Monica ; Neuman, Shlomo P ; Guadagnini, Alberto. / New scaling model for variables and increments with heavy-tailed distributions. In: Water Resources Research. 2015 ; Vol. 51, No. 6. pp. 4623-4634.
@article{cf9f1fa1a84b4a2d84bcdad9983b1972,
title = "New scaling model for variables and increments with heavy-tailed distributions",
abstract = "Many hydrological (as well as diverse earth, environmental, ecological, biological, physical, social, financial and other) variables, Y, exhibit frequency distributions that are difficult to reconcile with those of their spatial or temporal increments, ΔY. Whereas distributions of Y (or its logarithm) are at times slightly asymmetric with relatively mild peaks and tails, those of ΔY tend to be symmetric with peaks that grow sharper, and tails that become heavier, as the separation distance (lag) between pairs of Y values decreases. No statistical model known to us captures these behaviors of Y and ΔY in a unified and consistent manner. We propose a new, generalized sub-Gaussian model that does so. We derive analytical expressions for probability distribution functions (pdfs) of Y and ΔY as well as corresponding lead statistical moments. In our model the peak and tails of the ΔY pdf scale with lag in line with observed behavior. The model allows one to estimate, accurately and efficiently, all relevant parameters by analyzing jointly sample moments of Y and ΔY. We illustrate key features of our new model and method of inference on synthetically generated samples and neutron porosity data from a deep borehole. Key Points: A new statistical scaling model is developed, explored and applied Apparent inconsistency between variable and increment statistics is eliminated Parameters are estimated using variable and increment moments jointly",
keywords = "heavy-tailed distributions, neutron porosity, parameter estimation, scaling, sub-Gaussian",
author = "Monica Riva and Neuman, {Shlomo P} and Alberto Guadagnini",
year = "2015",
month = "6",
day = "1",
doi = "10.1002/2015WR016998",
language = "English (US)",
volume = "51",
pages = "4623--4634",
journal = "Water Resources Research",
issn = "0043-1397",
publisher = "American Geophysical Union",
number = "6",

}

TY - JOUR

T1 - New scaling model for variables and increments with heavy-tailed distributions

AU - Riva, Monica

AU - Neuman, Shlomo P

AU - Guadagnini, Alberto

PY - 2015/6/1

Y1 - 2015/6/1

N2 - Many hydrological (as well as diverse earth, environmental, ecological, biological, physical, social, financial and other) variables, Y, exhibit frequency distributions that are difficult to reconcile with those of their spatial or temporal increments, ΔY. Whereas distributions of Y (or its logarithm) are at times slightly asymmetric with relatively mild peaks and tails, those of ΔY tend to be symmetric with peaks that grow sharper, and tails that become heavier, as the separation distance (lag) between pairs of Y values decreases. No statistical model known to us captures these behaviors of Y and ΔY in a unified and consistent manner. We propose a new, generalized sub-Gaussian model that does so. We derive analytical expressions for probability distribution functions (pdfs) of Y and ΔY as well as corresponding lead statistical moments. In our model the peak and tails of the ΔY pdf scale with lag in line with observed behavior. The model allows one to estimate, accurately and efficiently, all relevant parameters by analyzing jointly sample moments of Y and ΔY. We illustrate key features of our new model and method of inference on synthetically generated samples and neutron porosity data from a deep borehole. Key Points: A new statistical scaling model is developed, explored and applied Apparent inconsistency between variable and increment statistics is eliminated Parameters are estimated using variable and increment moments jointly

AB - Many hydrological (as well as diverse earth, environmental, ecological, biological, physical, social, financial and other) variables, Y, exhibit frequency distributions that are difficult to reconcile with those of their spatial or temporal increments, ΔY. Whereas distributions of Y (or its logarithm) are at times slightly asymmetric with relatively mild peaks and tails, those of ΔY tend to be symmetric with peaks that grow sharper, and tails that become heavier, as the separation distance (lag) between pairs of Y values decreases. No statistical model known to us captures these behaviors of Y and ΔY in a unified and consistent manner. We propose a new, generalized sub-Gaussian model that does so. We derive analytical expressions for probability distribution functions (pdfs) of Y and ΔY as well as corresponding lead statistical moments. In our model the peak and tails of the ΔY pdf scale with lag in line with observed behavior. The model allows one to estimate, accurately and efficiently, all relevant parameters by analyzing jointly sample moments of Y and ΔY. We illustrate key features of our new model and method of inference on synthetically generated samples and neutron porosity data from a deep borehole. Key Points: A new statistical scaling model is developed, explored and applied Apparent inconsistency between variable and increment statistics is eliminated Parameters are estimated using variable and increment moments jointly

KW - heavy-tailed distributions

KW - neutron porosity

KW - parameter estimation

KW - scaling

KW - sub-Gaussian

UR - http://www.scopus.com/inward/record.url?scp=84937521622&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937521622&partnerID=8YFLogxK

U2 - 10.1002/2015WR016998

DO - 10.1002/2015WR016998

M3 - Article

AN - SCOPUS:84937521622

VL - 51

SP - 4623

EP - 4634

JO - Water Resources Research

JF - Water Resources Research

SN - 0043-1397

IS - 6

ER -